import pyforest
import warnings
warnings.filterwarnings('ignore')

df=pd.read_csv(r'D:\MTech\My_Projects\Bangalore_Traffic_Dataset\Banglore_traffic_Dataset.csv')

Inspecting the Data

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 8936 entries, 0 to 8935
Data columns (total 16 columns):
 #   Column                              Non-Null Count  Dtype  
---  ------                              --------------  -----  
 0   Date                                8936 non-null   object 
 1   Area Name                           8936 non-null   object 
 2   Road/Intersection Name              8936 non-null   object 
 3   Traffic Volume                      8936 non-null   int64  
 4   Average Speed                       8936 non-null   float64
 5   Travel Time Index                   8936 non-null   float64
 6   Congestion Level                    8936 non-null   float64
 7   Road Capacity Utilization           8936 non-null   float64
 8   Incident Reports                    8936 non-null   int64  
 9   Environmental Impact                8936 non-null   float64
 10  Public Transport Usage              8936 non-null   float64
 11  Traffic Signal Compliance           8936 non-null   float64
 12  Parking Usage                       8936 non-null   float64
 13  Pedestrian and Cyclist Count        8936 non-null   int64  
 14  Weather Conditions                  8936 non-null   object 
 15  Roadwork and Construction Activity  8936 non-null   object 
dtypes: float64(8), int64(3), object(5)
memory usage: 1.1+ MB

df.describe().transpose()

	count	mean	std	min	25%	50%	75%	max
Traffic Volume	8936.0	29236.048120	13001.808801	4233.000000	19413.000000	27600.000000	38058.500000	72039.000000
Average Speed	8936.0	39.447427	10.707244	20.000000	31.775825	39.199368	46.644517	89.790843
Travel Time Index	8936.0	1.375554	0.165319	1.000039	1.242459	1.500000	1.500000	1.500000
Congestion Level	8936.0	80.818041	23.533182	5.160279	64.292905	92.389018	100.000000	100.000000
Road Capacity Utilization	8936.0	92.029215	16.583341	18.739771	97.354990	100.000000	100.000000	100.000000
Incident Reports	8936.0	1.570389	1.420047	0.000000	0.000000	1.000000	2.000000	10.000000
Environmental Impact	8936.0	108.472096	26.003618	58.466000	88.826000	105.200000	126.117000	194.078000
Public Transport Usage	8936.0	45.086651	20.208460	10.006853	27.341191	45.170684	62.426485	79.979744
Traffic Signal Compliance	8936.0	79.950243	11.585006	60.003933	69.828270	79.992773	89.957358	99.993652
Parking Usage	8936.0	75.155597	14.409394	50.020411	62.545895	75.317610	87.518589	99.995049
Pedestrian and Cyclist Count	8936.0	114.533348	36.812573	66.000000	94.000000	102.000000	111.000000	243.000000

Traffic Data Column Profiling:

Traffic Volume = Count of vehicles crossing a specific point in a time interval.(Here in hours)

Average Speed refers to the mean velocity of all vehicles passing a given road segment or intersection during a specific time period (usually 1 hour).

The Travel Time Index (TTI) is a widely used traffic congestion indicator that compares the actual travel time experienced by drivers to the time it would take under ideal (free-flow) conditions.If it is 1.0 then no congestion(free-flow speed), 1.2 then 20% longer then free-flow, 1.5 then 50% longer than free flow or heavy congestion, 2.0 means twice, severe congestion.

Congestion Level is a measure of how crowded or slow traffic is on a road segment at a given time.(in percentage)

Road Capacity Utilization measures how fully a road is being used relative to its maximum designed capacity—that is, how close traffic volume is to what the road can handle efficiently.

Incident Reports refer to recorded events that disrupt normal traffic flow. These can include accidents, vehicle breakdowns, construction work, or any unexpected events on or near the road that may cause congestion, delays, or diversions.

Environmental Impact refers to how traffic conditions affect the environment.

Public Transport Usage refers to the number of people using buses, metros, or other public transit systems

Traffic Signal Compliance refers to how well drivers obey traffic signals—especially red lights—at intersections. It indicates driver behavior, traffic discipline, and intersection safety.

Parking Usage refers to how much of the available parking space is occupied at a given time and location. It reflects demand for parking and helps analyze traffic congestion, especially in commercial or dense areas like central Bengaluru.

Pedestrian and Cyclist Count refers to the number of people walking or cycling through a particular location during a specific time interval. It helps analyze non-motorized traffic flow, safety, infrastructure usage, and urban mobility trends.

df.columns

Index(['Date', 'Area Name', 'Road/Intersection Name', 'Traffic Volume',
       'Average Speed', 'Travel Time Index', 'Congestion Level',
       'Road Capacity Utilization', 'Incident Reports', 'Environmental Impact',
       'Public Transport Usage', 'Traffic Signal Compliance', 'Parking Usage',
       'Pedestrian and Cyclist Count', 'Weather Conditions',
       'Roadwork and Construction Activity'],
      dtype='object')

df.head()

	Date	Area Name	Road/Intersection Name	Traffic Volume	Average Speed	Travel Time Index	Congestion Level	Road Capacity Utilization	Incident Reports	Environmental Impact	Public Transport Usage	Traffic Signal Compliance	Parking Usage	Pedestrian and Cyclist Count	Weather Conditions	Roadwork and Construction Activity
0	2022-01-01	Indiranagar	100 Feet Road	50590	50.230299	1.500000	100.000000	100.000000	0	151.180	70.632330	84.044600	85.403629	111	Clear	No
1	2022-01-01	Indiranagar	CMH Road	30825	29.377125	1.500000	100.000000	100.000000	1	111.650	41.924899	91.407038	59.983689	100	Clear	No
2	2022-01-01	Whitefield	Marathahalli Bridge	7399	54.474398	1.039069	28.347994	36.396525	0	64.798	44.662384	61.375541	95.466020	189	Clear	No
3	2022-01-01	Koramangala	Sony World Junction	60874	43.817610	1.500000	100.000000	100.000000	1	171.748	32.773123	75.547092	63.567452	111	Clear	No
4	2022-01-01	Koramangala	Sarjapur Road	57292	41.116763	1.500000	100.000000	100.000000	3	164.584	35.092601	64.634762	93.155171	104	Clear	No

df.tail()

	Date	Area Name	Road/Intersection Name	Traffic Volume	Average Speed	Travel Time Index	Congestion Level	Road Capacity Utilization	Incident Reports	Environmental Impact	Public Transport Usage	Traffic Signal Compliance	Parking Usage	Pedestrian and Cyclist Count	Weather Conditions	Roadwork and Construction Activity
8931	2024-08-09	Electronic City	Hosur Road	11387	23.440276	1.262384	35.871483	57.354487	1	72.774	21.523289	83.530352	97.898279	211	Fog	No
8932	2024-08-09	M.G. Road	Trinity Circle	36477	45.168429	1.500000	100.000000	100.000000	3	122.954	29.822312	60.738488	60.355967	95	Clear	No
8933	2024-08-09	M.G. Road	Anil Kumble Circle	42822	22.028609	1.500000	100.000000	100.000000	1	135.644	43.185905	85.321627	61.333731	110	Clear	No
8934	2024-08-09	Jayanagar	South End Circle	20540	52.254798	1.020520	72.639152	97.845527	2	91.080	44.416043	89.586947	79.197198	94	Clear	No
8935	2024-08-09	Yeshwanthpur	Yeshwanthpur Circle	14705	31.128967	1.048720	43.409821	77.734621	1	79.410	26.616725	80.778753	60.602672	201	Rain	No

df.shape

(8936, 16)

Cleaning the Data

df.isnull().sum()

Date                                  0
Area Name                             0
Road/Intersection Name                0
Traffic Volume                        0
Average Speed                         0
Travel Time Index                     0
Congestion Level                      0
Road Capacity Utilization             0
Incident Reports                      0
Environmental Impact                  0
Public Transport Usage                0
Traffic Signal Compliance             0
Parking Usage                         0
Pedestrian and Cyclist Count          0
Weather Conditions                    0
Roadwork and Construction Activity    0
dtype: int64

df.duplicated().sum()

0

Since there are no null values and no duplicates in the dataset, I will proceed with Exploratory Data Analysis (EDA).

Exploratory Data Analysis (EDA)

import math
import plotly.graph_objects as go
from plotly.subplots import make_subplots


#select numeric columns
numeric_cols=df.select_dtypes(include='number').columns

#setup the grid
num_of_cols=len(numeric_cols)
cols_per_row=3
num_of_rows=math.ceil(num_of_cols/cols_per_row)
num_of_rows

# Setup subplot figure
fig = make_subplots(rows=num_of_rows, cols=cols_per_row, 
                    subplot_titles=numeric_cols)


for idx, col in enumerate(numeric_cols):
    row = idx // cols_per_row + 1
    col_pos = idx % cols_per_row + 1

    fig.add_trace(
        go.Histogram(x=df[col], name=col, nbinsx=20, marker_color='indianred'),
        row=row, col=col_pos
    )

# Update layout
fig.update_layout(
    height=300 * num_of_rows,
    width=1000,
    title_text="Histograms of Numerical Features",
    showlegend=True
)

fig.show()

Box Plot Analysis: Detecting Anomalies in Numeric Features

import math
import plotly.graph_objects as go
from plotly.subplots import make_subplots

# Select numeric columns
numeric_cols = df.select_dtypes(include='number').columns

# Setup the grid
cols_per_row = 3
num_of_cols = len(numeric_cols)
num_of_rows = math.ceil(num_of_cols / cols_per_row)

# Setup subplot figure
fig = make_subplots(rows=num_of_rows, cols=cols_per_row, 
                    subplot_titles=numeric_cols)

# Add a boxplot for each numeric column
for idx, col in enumerate(numeric_cols):
    row = idx // cols_per_row + 1
    col_pos = idx % cols_per_row + 1

    fig.add_trace(
        go.Box(y=df[col], name=col, marker_color='royalblue'),
        row=row, col=col_pos
    )

# Update layout
fig.update_layout(
    height=300 * num_of_rows,
    width=1000,
    title_text="Boxplots of Numerical Features",
    showlegend=False
)

fig.show()

Correlation Matrix: Identifying Relationships Between Numeric Features

# Compute correlation
numeric_df = df.select_dtypes(include='number')
corr = numeric_df.corr()

# Mask the upper triangle (optional, to reduce redundancy)
mask = np.triu(np.ones_like(corr, dtype=bool))

# Set up the color palette
cmap = sns.diverging_palette(250, 10, as_cmap=True)

# Plot
plt.figure(figsize=(12, 10))
sns.heatmap(corr,mask=mask, cmap=cmap, annot=True, fmt=".2f",
            linewidths=0.6, linecolor='gray', square=True, cbar_kws={"shrink": 0.8})
plt.title("Correlation Matrix of Dataset Variables", fontsize=16, pad=15)
plt.xticks(rotation=45, ha='right', fontsize=10)
plt.yticks(rotation=0, fontsize=10)
plt.tight_layout()
plt.show()

The distribution of total traffic volume among different junctions across areas.

heatmap_data = df.pivot_table(values='Traffic Volume', index='Area Name', columns='Road/Intersection Name', aggfunc='sum')
heatmap_data
fig = px.imshow(
    heatmap_data,
    labels=dict(x="Road/Intersection Name", y="Area Name", color="Traffic Volume"),
    title="Heatmap of Traffic Volume Across Areas and Roads",
    text_auto=False,
    width=1000,  
    height=600,

)
fig.show()

df.groupby(['Area Name','Road/Intersection Name'])['Traffic Volume'].agg('sum').sort_values(ascending=False)

Area Name        Road/Intersection Name
Koramangala      Sony World Junction       28366028
Indiranagar      CMH Road                  28046255
                 100 Feet Road             27482893
Koramangala      Sarjapur Road             27329166
M.G. Road        Anil Kumble Circle        26756138
                 Trinity Circle            26229802
Jayanagar        South End Circle          14649543
                 Jayanagar 4th Block       14207817
Hebbal           Hebbal Flyover            12700794
                 Ballari Road              12505777
Whitefield       Marathahalli Bridge       10220293
                 ITPL Main Road             9839717
Yeshwanthpur     Yeshwanthpur Circle        7190364
                 Tumkur Road                6705391
Electronic City  Hosur Road                 4586351
                 Silk Board Junction        4436997
Name: Traffic Volume, dtype: int64

df.groupby(['Area Name','Road/Intersection Name'])['Traffic Volume'].agg('mean').round(2).sort_values(ascending=False)

Area Name        Road/Intersection Name
Koramangala      Sony World Junction       41470.80
                 Sarjapur Road             40189.95
M.G. Road        Trinity Circle            35350.14
                 Anil Kumble Circle        35251.83
Indiranagar      CMH Road                  32611.92
                 100 Feet Road             31956.85
Hebbal           Hebbal Flyover            26794.92
                 Ballari Road              26272.64
Jayanagar        South End Circle          24704.12
                 Jayanagar 4th Block       24496.24
Whitefield       ITPL Main Road            21866.04
                 Marathahalli Bridge       20772.95
Yeshwanthpur     Yeshwanthpur Circle       19277.12
                 Tumkur Road               18574.49
Electronic City  Hosur Road                16557.22
                 Silk Board Junction       16134.53
Name: Traffic Volume, dtype: float64

Traffic Volume Distribution by Area and Junction in percentage

import plotly.express as px

# Step 1: Calculate average traffic volume
avg_volume = df.groupby(['Area Name', 'Road/Intersection Name'])['Traffic Volume'].mean().reset_index()

# Step 2: Sunburst chart with percentage inside each area
fig = px.sunburst(
    avg_volume,
    path=['Area Name', 'Road/Intersection Name'],   # Hierarchical levels
    values='Traffic Volume',                        # Size of each slice
    color='Traffic Volume',                         # Optional: use color to represent volume
    color_continuous_scale='RdYlGn_r',              # Color scale: green (low) to red (high)
    title='Average Traffic Volume by Area and Junction',
    width=1200,
    height=800
    
)

# Step 3: Update hover to show percentage share
fig.update_traces(insidetextorientation='radial', textinfo='label+percent entry')

fig.show()

Assessing the Environmental Impact of Traffic Volume

import seaborn as sns
import matplotlib.pyplot as plt

plt.figure(figsize=(8,6))
sns.scatterplot(data=df, x='Environmental Impact', y='Traffic Volume', hue='Congestion Level', palette='coolwarm', alpha=0.7)
plt.title('Traffic Volume vs Environmental Impact')
plt.ylabel('Environmental Impact')
plt.xlabel('Traffic Volume')
plt.show()

As traffic volume increases across key Bengaluru intersections, environmental stress also rises significantly. High vehicle density leads to:

Increased air pollution: More vehicles emit higher levels of pollutants such as PM2.5, NO₂, and CO₂.

Elevated PM2.5 levels: Traffic-heavy areas consistently show higher concentrations of fine particulate matter, especially during peak hours.

Fuel inefficiency and GHG emissions: Stop-and-go traffic leads to more fuel combustion, contributing to greenhouse gases.

Noise pollution: Heavily congested junctions see a spike in noise levels, impacting human health.

Heat generation: Dense traffic and reduced greenery contribute to urban heat islands, raising local temperatures.

Variation in Traffic Volume Across Weekdays and Weekends

# Ensure your column is in datetime format
df['Date'] = pd.to_datetime(df['Date'])

# Create a new column for day of the week (0=Monday, 6=Sunday)
df['DayOfWeek'] = df['Date'].dt.dayofweek

# Label as Weekday or Weekend
df['WeekType'] = df['DayOfWeek'].apply(lambda x: 'Weekend' if x >= 5 else 'Weekday')

#cross check the dates
df[['Date', 'DayOfWeek', 'WeekType']].sample(10)

	Date	DayOfWeek	WeekType
8655	2024-07-12	4	Weekday
1996	2022-07-31	6	Weekend
4165	2023-03-22	2	Weekday
6357	2023-11-10	4	Weekday
1113	2022-04-29	4	Weekday
8451	2024-06-20	3	Weekday
7157	2024-02-02	4	Weekday
3339	2022-12-22	3	Weekday
1099	2022-04-27	2	Weekday
1913	2022-07-23	5	Weekend

# Get average and total traffic volume by WeekType
traffic_comparison = df.groupby('WeekType')['Traffic Volume'].agg(['mean', 'sum', 'count']).round(2).reset_index()
print(traffic_comparison)

  WeekType      mean        sum  count
0  Weekday  29297.17  186740164   6374
1  Weekend  29083.98   74513162   2562

import plotly.express as px

fig = px.scatter(
    df,
    x='Traffic Volume',
    y='Average Speed',
    color='WeekType',
    facet_col='WeekType',
    title='Traffic Volume vs Average Speed: Weekday vs Weekend',
    opacity=0.7,
    height=500,
    width=1000
)

fig.update_layout(showlegend=False)
fig.show()

Our scatter plot analysis comparing "Traffic Volume vs. Average Speed: Weekday vs Weekend" clearly illustrates that weekdays are significantly more crowded and congested than weekends. The weekday plot (red dots) shows a much denser concentration of data points, particularly at higher Traffic Volumes leading to consistently lower Average Speeds. This stark contrast with the weekend plot (blue dots), which displays generally lower traffic volumes and less severe speed reductions, provides a strong basis for strategic planning.

Therefore, a key recommendation for scheduling disruptive urban infrastructure projects, such as flyover construction or major road repair work, is to prioritize and conduct these activities predominantly during weekend periods.

Traffic Volume Under Varying Weather Conditions

df['Weather Conditions'].unique()

array(['Clear', 'Overcast', 'Fog', 'Rain', 'Windy'], dtype=object)

import plotly.express as px

# Ensure the column is clean
df['Weather Conditions'] = df['Weather Conditions'].astype(str).str.strip().str.title()

# Group by weather condition
df_grouped = df.groupby('Weather Conditions', as_index=False)['Traffic Volume'].sum()

# Create the chart
fig = px.bar(
    df_grouped.sort_values('Traffic Volume'),
    x='Weather Conditions',
    y='Traffic Volume',
    color='Weather Conditions',
    color_discrete_sequence=px.colors.qualitative.Prism,
    title='Traffic Volume by Weather Conditions',
    height=600,
    width=1000
)

# Layout styling
fig.update_layout(
    title_font=dict(size=24, family='Arial', color='black'),
    xaxis=dict(
        title='Weather Conditions',
        titlefont=dict(size=18),
        tickfont=dict(size=14),
        showgrid=True,
        gridcolor='#e0e0e0',
        linecolor='black',
        linewidth=1
    ),
    yaxis=dict(
        title='Traffic Volume',
        titlefont=dict(size=18),
        tickfont=dict(size=14),
        showgrid=True,
        gridcolor='#ffe0e0',
        linecolor='black',
        linewidth=1
    ),
    plot_bgcolor='white',
    paper_bgcolor='white',
    margin=dict(l=60, r=40, t=80, b=60),
    showlegend=False
)

# Add border to bars
fig.update_traces(
    marker_line_color='black',
    marker_line_width=1.5,
    opacity=0.9
)

fig.show()

Analysis of Traffic Volume by Weather Conditions

The bar chart illustrates the significant impact of various weather conditions on traffic volume.

Key Findings:

Dominance of Clear Weather: Clear weather consistently registers the highest traffic volume, significantly outstripping all other conditions. Its volume is approximately four times greater than the next highest category (overcast conditions).

Reduced Traffic in Other Conditions: Traffic volume substantially decreases under overcast, fog, rain, and windy conditions. Among these, windy weather consistently shows the lowest traffic volume.

Conclusion:

This analysis strongly suggests a direct correlation between clear weather and high traffic density, indicating that favorable weather is a primary factor influencing road usage and congestion.

Analyzing the Relationship Between Construction Activity and Traffic Congestion

df['Roadwork and Construction Activity'].unique()

array(['No', 'Yes'], dtype=object)

df.groupby('Roadwork and Construction Activity')[['Traffic Volume', 'Travel Time Index']].mean().reset_index()

	Roadwork and Construction Activity	Traffic Volume	Travel Time Index
0	No	29194.610504	1.375513
1	Yes	29614.436508	1.375922

no_work = df[df['Roadwork and Construction Activity'] == 'No']
yes_work = df[df['Roadwork and Construction Activity'] == 'Yes']

tti_increase = ((yes_work['Travel Time Index'].mean() - no_work['Travel Time Index'].mean()) /
                 no_work['Travel Time Index'].mean()) * 100

volume_increase = ((yes_work['Traffic Volume'].mean() - no_work['Traffic Volume'].mean()) /
                    no_work['Traffic Volume'].mean()) * 100

print(f"TTI increases by {tti_increase:.2f}% due to roadwork.")
print(f"Traffic Volume increases by {volume_increase:.2f}% due to roadwork.")

TTI increases by 0.03% due to roadwork.
Traffic Volume increases by 1.44% due to roadwork.

import plotly.graph_objects as go

fig = go.Figure(data=[
    go.Bar(name='Without Roadwork', x=['Traffic Volume', 'Travel Time Index'],
           y=[no_work['Traffic Volume'].mean(), no_work['Travel Time Index'].mean()]),
    go.Bar(name='With Roadwork', x=['Traffic Volume', 'Travel Time Index'],
           y=[yes_work['Traffic Volume'].mean(), yes_work['Travel Time Index'].mean()])
])

fig.update_layout(
    barmode='group',
    title='Impact of Roadwork on Congestion Levels',
    yaxis_title='Average Value',
    plot_bgcolor='white'
)

fig.show()

The analysis shows that the Travel Time Index (TTI) increases by only 0.03% in the presence of roadwork or construction activity. This minimal rise suggests that roadwork has a negligible impact on overall congestion levels in the observed areas. Therefore, other factors like traffic volume, weather, or time of day likely play a more significant role in affecting travel delays.

Statistical Summary and Inference from Traffic Dataset

Does traffic volume differ significantly between weekdays and weekends?

#convert into date time from object
df['Date']=pd.to_datetime(df['Date'])

df['Day_of_week']=df['Date'].dt.dayofweek

df['WeekType']=df['Day_of_week'].apply(lambda x: 'weekend' if x>=5 else 'weekday')

weekday_volume=df[df['WeekType'] == 'weekday']['Traffic Volume']
weekend_volume=df[df['WeekType'] == 'weekend']['Traffic Volume']

Traffic Volume is a continuous (numeric) variable.

Day Type (Weekday vs Weekend) is a categorical variable with two independent groups.

Each observation belongs exclusively to one group (i.e., no overlap between weekday and weekend data).

Therefore, an independent t-test was used to compare the mean traffic volume across these two groups and assess whether the observed differences are statistically significant.

from scipy.stats import ttest_ind
t_stat,p_value=ttest_ind(weekday_volume,weekend_volume,equal_var=False)

print(f"T-statistics : {t_stat:.2f}")
print(f"P-Value: {p_value:.2f}")

T-statistics : 0.70
P-Value: 0.48

Assumption

H0: There is no difference in traffic volumes between weekdays and weekends.

H1: There is a difference in traffic volumes between weekdays and weekends.

Result: The p-value = 0.48 is greater than the significance level (0.05), so we fail to reject the null hypothesis (H0).

Conclusion: There is no statistically significant difference in traffic volumes between weekdays and weekends.

Does increased parking usage have a significant impact on congestion levels

To evaluate the relationship between parking usage (independent numeric variable) and congestion level (dependent numeric variable, measured by the Travel Time Index), we employed the Pearson correlation test. This method is appropriate for assessing the strength and direction of a linear association between two continuous variables.

from scipy.stats import pearsonr

corr,p_value=pearsonr(df['Parking Usage'],df['Travel Time Index'])

print(f"The correlation :{corr:.2f}")
print(f"P-Value :{p_value:.2f}")

The correlation :-0.02
P-Value :0.03

Assumption:

H0: Parking usage has no significant impact on congestion levels.

H1: Parking usage has a significant impact on congestion levels.

Since the p-value = 0.03 is less than the significance level (0.05), we reject the null hypothesis (H₀).

Conclusion: There is a statistically significant relationship between parking usage and congestion levels, indicating that parking usage has a significant impact on congestion.

Is there a significant relationship between weather conditions and the frequency of incident reports

One-Way ANOVA is the appropriate test in this case because the independent variable (weather conditions) is categorical with more than two levels, and the dependent variable (incident reports) is numeric. The goal is to determine whether the mean number of incident reports differs significantly across various weather conditions.

weather_conditions=df.groupby('Weather Conditions')['Incident Reports'].agg('mean').sort_values(ascending=False)

weather_conditions

Weather Conditions
Fog         1.617310
Windy       1.593458
Overcast    1.577160
Clear       1.561555
Rain        1.551391
Name: Incident Reports, dtype: float64

from scipy.stats import f_oneway
groups = [group['Incident Reports'].values 
          for name, group in df.groupby('Weather Conditions')]

f_stat,p_value=f_oneway(*groups)

print(f_stat)
print(p_value)

0.3867433727672072
0.8182823936319665

Assumption

H0: There is no significant difference in the average number of incident reports across different weather conditions.

H1:There is significant difference in the average number of incident reports across different weather conditions.

Since your p-value = 0.818 is greater than 0.05, we fail to reject the null hypothesis.

There is no statistically significant difference in the mean number of incident reports across the different weather conditions. In other words, weather conditions do not significantly influence the frequency of incident reports — based on this dataset.

df.groupby('Weather Conditions')['Incident Reports'].sum().sort_values(ascending=False)

Weather Conditions
Clear       8473
Overcast    2044
Fog         1551
Rain        1283
Windy        682
Name: Incident Reports, dtype: int64

Incident reports accoding to weather conditions

incident_counts = df.groupby('Weather Conditions')['Incident Reports'].sum()
weather_counts = df['Weather Conditions'].value_counts()

incident_rate = (incident_counts / weather_counts) * 1000 # per 1000 records
incident_rate.sort_values(ascending=False)

Weather Conditions
Fog         1617.309698
Windy       1593.457944
Overcast    1577.160494
Clear       1561.555474
Rain        1551.390568
dtype: float64

Does traffic volume have a significant impact on environmental impact?

corr,p_value=pearsonr(df['Traffic Volume'],df['Environmental Impact'])

print(f"The correlation value: {corr:.2f}")
print(f"The P-value: {p_value:.2f}")

The correlation value: 1.00
The P-value: 0.00

Assumption:

H0: There is no correlation between traffic volume and environmental impact.

H1:There is a significant correlation between traffic volume and environmental impact.

The Pearson correlation coefficient is 1.00, indicating a perfect positive linear relationship between traffic volume and environmental impact.

The p-value is 0.00, which is less than 0.05, so we reject the null hypothesis.

Conclusion: There is a statistically significant and perfect positive correlation between traffic volume and environmental impact — as traffic volume increases, environmental impact increases proportionally

Does public transport usage have a significant impact on traffic volume?

corr, p_value = pearsonr(df['Public Transport Usage'], df['Traffic Volume'])
print(f"The correlation value: {corr:.2f}")
print(f"The P-value: {p_value:.2f}")

The correlation value: -0.00
The P-value: 0.91

Assumption:

H0: Public transport uasge has no impact on traffic volume.

H1: Public transport usage significantly affects traffic volume

Since the p-value is 0.91, which is much greater than 0.05, we fail to reject the null hypothesis.

pwd

'C:\\Users\\Admin\\M_TECH\\My_Projects'